388 



FIGURE 3. Schematic diagram for nuclei measurements. 



detected as an absolute value of the root mean 

 square . 



Relative Measurement of Cavitation Nuclei 



Size distributions of gas nuclei are measured by 

 using the sound-attenuation method of Schiebe 1969. 

 The measuring system is shown in Figure 3 . The 

 frequency range of swept pulses was 20kHz- 1000 kHz. 

 Both probes for emission and reception were 25mm 

 diameters, made of a crystal, and exposed directly 

 to water. The measurements were relative ones for 

 comparison between the three cases of no grid and 

 grids No. 1 and No. 2 because the system has not 

 yet been calibrated for bubbles with prescribed 

 definite diameters. 



Measurements were carried out at four positions 

 in the spanwise direction at the mid-chord of 

 hydrofoil perpendicular to the free stream and the 

 hydrofoil span. 



Observations and Measurements of Cavitation 



Cavitation inceptions are seen by the naked eye 

 tinder 50Hz, stroboscopic 3vis flash illumination. 

 An incipient cavitation number is defined by using 

 the static pressure at which the inception is 

 detected while reducing the static pressure at a 

 low rate and the local free stream velocity. How- 

 ever, in the boundary layers the velocities at 

 outside edges are taken while the free stream 

 velocity is kept at the prescribed value. Desin- 

 ences are too intermittent and indefinite to be 

 detected definitely in the course of raising the 

 static pressure. 



For the measurements of positions of inception 

 and the observations of appearances of cavitation 

 biibbles or cavities, photographs of 3 ps exposure 

 and high-speed motion pictures of 3000 frames per 

 second and 2 ys exposure for each frame were taken. 

 For the high-speed photography, the high-speed 

 camera, FASTAX, was used synchronized with the 

 high-speed stroboscope made by E. G. and E Co. Ltd. 

 For the measurements of average locations and shapes 

 of cavitation regions, photographs of 1/60 s 

 exposure were used. 



3. RESULTS OF EXPERIMENT AND DISCUSSIONS 

 Shear Flow at Measuring Section 



The velocity profiles normalized by each velocity 

 at the mid-span and the distributions of the static 

 pressure expressed as the difference from one at 

 the side wall and normalized by each dynamic pressure 

 at the mid- span for the grids No. 1 and No. 2 are 

 shown in Figure 4. The flow shear for grid No. 1 

 is uniform in the free stream core and the non- 

 dimensional shear factor is 0.15. That for grid 

 No. 2 is about the same as for grid No. 1 at half 

 the core of the free stream on the high-speed side 

 but smaller at the other half. The non-dimensional 

 shear factor is 0.06. Both have boundary layers of 

 10^ thickness span on both sides. The static 

 pressure is higher in the free stream core than on 

 the side walls by about Vfo or a little more of the 

 dynamic pressure at mid-span. Scatters of plots 

 are within the accuracy of this experiment. 



Spanwise Distribution of Turbulence 



00 01 0.2 03 0,4 05 06 07 08 09 10 



(a) Grid No.l 



001 

 



00 01 0.2 03 04 05 06 07 08 09 10 



y/h 



(b) Grid No. 2 



FIGURE 4. Velocity and static pressure distribution 

 for shear grids . 



Root mean squares of two components of turbulent 

 velocity, one stream-wise and the other perpendic- 

 ular to it ahd the hydrofoil span, are measured 

 in every free stream, and shown in Figure 5 

 normalized by Uc. The velocity at the mid-span was 

 kept at 9.86 m/s . When both are expressed as the 

 turbulence levels based on the local velocity of 

 free stream, U, for the cases of the two shear grids, 

 both u'/U and w'/U vary so little in the spanwise 

 direction that they can be regarded as constant 



10.0 



- 80 



■60 



4.0 



20 



FIGURE 5. Spanwise 

 distribution of 

 turbulence. 



Uc = 9.86m/s 





\ \ 





0.2 



0.4 0.6 



08 10 

 y/h 



